Advanced Dynamic Financial Modeling

Advanced Dynamic Financial Modeling

Carl Seidman

Seidman Financial

CPA/CFF, CIRA, CFE, CGMA, AM

How does this resonate with you?

Why Many Financial Models Fail

Today’s Model Requirements The Model ‘Evolution’

▪ Specific objectives and usage

▪ Limited maintenance

▪ Contained data

▪ Limited time periods

▪ Generally requires:

o Less-frequent updates

o Less flexibility

o Fewer dynamics

▪ Broad scope and use

▪ Ongoing maintenance

▪ Growing data sets

▪ Evolving time periods

▪ Generally requires:

o More-frequent updates

o More flexibility

o Greater dynamics

Excel is an extremely

flexible tool however…

Intentional

Scale

Incorporates

New

Data Easily

Manual

Involvement

Seamless

Agile

Updates

Intentional

Scale

Incorporates

New

Data Easily

Seamless

Agile

Updates

Manual

Involvement

Dynamic Modeling

Building robust yet flexible

models that:

✓ Scales intentionally

✓ Seamlessly agile updates

✓ Incorporates new data easily

✓ Minimizes manual

involvement



Intentional

Scale

Incorporates

New

Data Easily

Manual

Involvement

Seamless

Agile

Updates

Intentional

Scale

Incorporates

New

Data Easily

Seamless

Agile

Updates

Manual

Involvement

Dynamic Modeling


models that can




Intentional

Scale

Incorporates

New

Data Easily

Manual

Involvement

Seamless

Agile

Updates

Intentional

Scale

Incorporates

New

Data Easily

Seamless

Agile

Updates

Manual

Involvement

Dynamic Modeling


models that:





Intentional

Scale

Incorporates

New

Data Easily

Manual

Involvement

Seamless

Agile

Updates

Intentional

Scale

Incorporates

New

Data Easily

Seamless

Agile

Updates

Manual

Involvement

Dynamic Modeling


models that:






Intentional

Scale

Incorporates

New

Data Easily

Manual

Involvement

Seamless

Agile

Updates

Intentional

Scale

Incorporates

New

Data Easily

Seamless

Agile

Updates

Manual

Involvement

Dynamic Modeling


models that:





involvement

The objective of financial

modeling isn’t building a model

The objective is to provide

end-users great intelligence

So they can make better decisions

How might the

model need to

change?

What new data is

likely to enter

the model?

What degree of

dynamics/flexibility

is necessary?

How might the

data need to be

presented?(or is too much)?

ExcelSaaS

Platform

How Dynamic Functionality Differs

▪ OFFSET(

▪ INDIRECT(

▪ INDEX( + MATCH(

▪ SEQUENCE(

▪ UNIQUE(

▪ FILTER(

▪ SORT(

▪ SORTBY(

▪ RANDARRAY(

▪ DYNAMIC (#) ARRAYS

▪ Dynamic Tables

▪ Dynamic References

▪ Dynamic Ranges

▪ Dynamic Arrays

Excel Dynamic Functionality

DYNAMIC TABLES

Allows for the dynamic referencing

of expanding and contracting data sets

without requiring a formula update

Dynamic

Tables


✓ Seamlessly agile

updates

✓ Incorporates new

data easily


involvement

Dynamic

Tables✓ Tables can be named

and referenced

✓ Entire data sets and

columns can be

referenced by a

single formula

✓ Formulas reference

dynamic ranges

OFFSET FUNCTION

Allows for the dynamic referencing

of expanding and contracting data sets

without requiring a formula updates

Why You Should Use It:

OFFSET FUNCTION

OFFSET(reference, rows, cols, [height], [width])

Reference: The reference from

which you want to base the offset.

Rows: The number of rows, up or

down, that you want the upper-left

cell to refer to.

Cols: The number of columns, to the

left or right, that you want the upper-

left cell of the result to refer to.

Source: Microsoft

Height: The height, in number of rows,

that you want the returned reference to

be. Height must be a positive number.

Width: The width, in number of

columns, that you want the returned

reference to be. Width must be a

positive number.

Microsoft is incorrect!

It can ALSO be negative

OFFSET FUNCTION

OFFSET FUNCTIONWidth = 7

Height = 9

Make Dynamic!

Make Dynamic!

OFFSET FUNCTIONWidth = 7

Height = 9

=OFFSET(reference, rows, cols, [height], [width])

=OFFSET( E2 , 0 , 0 , 9 , 7 )

OFFSET FUNCTION

= OFFSET ( E2, 0 , 0 , 9 , 7 )

Height should expand/contract

up and down

as data added/removed

Use:

COUNT(

MIN(

MAX(

ISNUM(

ISBLANK(

MATCH(

to determine the size and shape

of the data set

OFFSET FUNCTION

= OFFSET ( E2, 0 , 0 , 9 , 7 )

Width should expand/contract

left and right

as data added/removed

OFFSET FUNCTION

= OFFSET ( E2, 0 , 0 , 9 , 7 )

This formula should be named

(Ex: DYNAMIC_VEHICLE_DATA)

And can be dynamically referenced EVERYWHERE in the model

When To Use Dynamic Tables vs. OFFSET

Dynamic Tables▪ When data is symmetrical

and complete with no skipped lines or gaps

▪ When entire columns will be referenced

▪ When data is expected to be appended at the bottom

▪ When there is no data series

OFFSET Function▪ When data is asymmetrical

and may have skipped lines and gaps

▪ When entire columns and/or rows will be referenced

▪ When data is expected to be added to the right or left

▪ When there is a data series

DYNAMIC INDEX RANGES

Allows for the dynamic referencing of

expanding and contracting data ranges

prompted by the user’s input

Why You Should Use Them:

INDEX

INDEX finds a

corresponding value in a

table or a range by

coordinate

MATCH

INDEX(array, row_index_num, col_index_num)

MATCH finds a

corresponding location

in a table or a range

MATCH(lookup_value, lookup_array, [type])

Dynamic INDEX MATCH Ranges

Total


Dynamic Header

Dynamic SUM

Total


="Active Scenario: "&Assumptions!$C$3

=SUM(INDEX(D12:I12,1,MATCH(M3,D4:I4,0)):INDEX(D12:I12,1,MATCH(M4,D4:I4,0)))

Total

INDIRECT FUNCTION

Allows for the reference of data, cells

and worksheets without requiring a

physical link

Why You Should Use It:

Physical Link

Returns the contents of

a cell based upon a

physical formula

INDIRECT

Returns the contents of

a cell based upon a text

string reference

INDIRECT(ref_text)

INDIRECT FUNCTION

Text String

Physical link

Dynamic INDIRECT

DYNAMIC ARRAYS

Allows for various expansion of arrays

and spills of dynamic data and ranges

Why You Should Use Them:

Introduced in Summer 2020

✓ Returns arrays (or ranges) of

results rather than a single

result

✓ Accomplishes tasks that once

required extremely complex

or nuanced formulas

ExamplesDynamic Arrays

SEQUENCE: Generates a sequence of

numbers in a range of rows and/or columns

UNIQUE: extracts unique items from a range

of cells/table

FILTER: Filters a range of cells/table based

on input criteria

SORT: Sorts a list by a specified other

column in that list

SORTBY: Sorts a list by a second list

RANDARRAY: Generates a range of random

numbers in an array

Dynamic Arrays

Can we build formulas that allow this model to expand and

update automatically with minimal manual involvement?

SEQUENCE FUNCTION

Rows: the number of rows to return

Columns: the number of columns to return

Start: the first number in the sequence

Step: the amount to increment each subsequent

value in the sequence

SEQUENCE ( rows , [columns] , [start] , [step] )

SEQUENCE FUNCTION

SEQUENCE ( rows , [columns] , [start] , [step] )

Dynamic (#) Arrays

Formulas with a hash (#) mark allow for expansion of an

array driven by a precedent array

One Final Thought

Doesn’t dynamic functionality

increase complexity for end-users?

Remember: The objective of financial

modeling isn’t building a model

It’s about providing end-users

good information

So they can make better decisions

Don’t avoid great functionality

because it’s novel for end-users

Don’t make models so complex

that they lose the end-users

Intentionally

Scale

Incorporate

New

Data Easily

Minimize

Manual

Involvement

Update

Seamlessly

DO develop your

model-builders and end-users to:

Contact and Support

Carl Seidman

[email protected]

www.seidmanfinancial.com

Documents

Advanced Dynamic Financial Modeling